PSRA: A data model for managing data in sensor networks

Haiyang Liu, San Yih Hwang, Jaideep Srivastava

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

Sensor data streams exhibit unique characteristics such as inherent information uncertainty, intrinsic data sample correlations (both within and across streams) and resource constraints. In this paper, we introduce a new data model, called Probabilistic Stream Relational Algebra (PSRA), that extends conventional relational model to capture these new characteristics faced in managing data in sensor networks. New data types, new operations and essential strategies are incorporated into PSRA to facilitate flexible data modeling and resource-efficient operations. We show that operators in PSRA are non-blocking and more expressive than conventional relational model and existing deterministic data stream processing models. A demonstrating application implementing key operations in PSRA is provided to show the advantages of utilizing PSRA in managing data in sensor networks.

Original languageEnglish (US)
Title of host publicationProceedings - Thirteenth International Symposium on Temporal Representation and Reasoning, TIME 2006
Pages540-547
Number of pages8
Volume2006 II
DOIs
StatePublished - Dec 15 2006
EventIEEE International Conference on Sensor Networks, Ubiquitous, and Trustworthy Computing - Taichung, Taiwan, Province of China
Duration: Jun 5 2006Jun 7 2006

Other

OtherIEEE International Conference on Sensor Networks, Ubiquitous, and Trustworthy Computing
CountryTaiwan, Province of China
CityTaichung
Period6/5/066/7/06

Fingerprint Dive into the research topics of 'PSRA: A data model for managing data in sensor networks'. Together they form a unique fingerprint.

Cite this